Appen H2 Earnings Call Highlights

Appen (ASX:APX) executives said FY25 marked a “year of meaningful progress,” pointing to revenue growth excluding the impact of Google, improving margins driven by higher-value generative AI work, and particularly strong performance from the company’s China segment.

FY25 results: revenue growth, margin improvement, and strong Q4 finish

Management reported group revenue of $230.8 million, up 4.5% year-over-year excluding the impact of Google. The company said growth was driven predominantly by new project wins and expansions in generative AI, which executives described as a “clear growth driver” and an indicator the firm is executing against its strategy.

On profitability, Appen delivered $12.2 million in underlying EBITDA (excluding FX), with a full-year underlying EBITDA margin of 5.3%. The company highlighted a strong finish to the year, with Q4 underlying EBITDA margin of 18.2%. Gross margin improved 100 basis points to 40.3%, which management attributed to a greater mix of generative AI projects and wins in “higher value” work.

The company also emphasized operational efficiencies, noting it realized $10 million in annualized cost efficiencies in FY25 through technology, innovation, and automation. Finance leadership added that these efficiencies were net of talent upgrades.

Segment performance: Appen China surges while Appen Global rebounds in Q4

Appen reported sharply different trends across its two segments in FY25.

  • Appen Global: Full-year revenue was $127.9 million, down 21.1% year-over-year (excluding Google). The company said the year was impacted by lower volumes than expected in Q1 to Q3, but it ended strongly. In Q4, Appen Global revenue was $41.4 million, up 56% versus Q3, and Q4 EBITDA was $10.2 million at a 24.6% margin. Executives attributed the Q4 acceleration to new project wins, including a $10 million-plus generative AI opportunity that grew faster than anticipated and carried into FY26.
  • Appen China: Revenue rose 74.8% to $102.9 million, while EBITDA increased 640% to $10.6 million. Management said growth was driven by new and expanding LLM-related projects, as well as increasing revenue from high-margin pre-built datasets. The company also cited scaling benefits from “tight OpEx controls” as revenue expanded, and said Appen China exited the year with annualized revenue exceeding $135 million.

Across the group, Appen said 44.1% of Q4 revenue came from generative AI, up from 34.8% in the prior-year period.

Cash flow and accounting items

Appen ended the year with $59.8 million in cash, up $5 million from December 2024 (the company also provided an Australian dollar equivalent of AUD 89.5 million). Cash flow from operations improved by $23.4 million to $22.4 million.

Management noted operating cash flow benefited from the timing of a customer payment received in early January instead of late December as scheduled, but added that even adjusting for that timing effect, the business still delivered “approximately 100% conversion of EBITDA to cash flow from operations.”

On earnings, finance leadership said underlying NPAT improvement was “minimal” despite the EBITDA increase due to higher non-cash amortization. Statutory NPAT was additionally impacted by a $5 million acceleration of non-cash amortization related to acquired platforms.

Strategy and market drivers: globalization, enterprise AI, and robotics

In outlining the market opportunity, management pointed to three structural trends driving demand for Appen’s services:

  • Globalization of consumer AI, which the company linked to the digital advertising market and the need for multilingual, culturally nuanced human data as AI products scale internationally.
  • Enterprise AI adoption, including momentum in agentic AI. Management said model builders are focusing heavily on agentic systems and that Appen is “in the room” with leading AI labs as they experiment, with expectations this work will evolve into a major growth driver.
  • New form factors, particularly humanoid and industrial robotics, where models require real-world grounded data. Appen said it is already active in robotic simulation and real-world data collection projects.

Executives also emphasized investments in talent, stating Appen added more than 20 industry experts across go-to-market, operations, and workforce management over the past year. The company said these hires are enabling more technical, consultative work and helping it participate earlier in experimental projects that can scale quickly.

FY26 priorities and guidance

Looking ahead, management outlined four FY26 priorities: data quality, customer growth focused on hyperscalers and foundation model builders, new data segments through co-innovation with customers, and continued operational efficiency building on FY25 cost reductions.

Appen also issued FY26 group guidance, citing improved confidence partly because China has become a larger portion of revenue and is “less variable” quarter-to-quarter than project-based work. The company guided to:

  • Revenue of $270 million to $300 million
  • Underlying EBITDA (before FX) margin of 5% to 10%

On seasonality, management said Appen Global historically skews toward a stronger second half, noting that January and February can feature customer replanning for those operating on calendar-year cycles. When asked about synthetic data and automation, executives said synthetic data is an important part of the market and is being incorporated into solutions alongside human data to deliver quality and unit economics.

In closing remarks, leadership reiterated confidence in the AI data market and said the company is “winning in the right parts of the market” as it enters FY26 with strong Q4 momentum and a balance sheet it believes provides flexibility to execute its strategy.

About Appen (ASX:APX)

Appen Limited operates as an AI lifecycle company that provides data sourcing, data annotation, and model evaluation solutions in Australia, the United States, and internationally. It operates through two segments, Global Services and New Markets. The company provides a platform for the AI data development process. Its platform provides data annotation, such as workflow management and collaboration, data organization, communication, and performance analysis; audio transcription; data collection and categorization; data monitoring; and dataset creation, model performance evaluation, fine-tuning, A/B testing for validation.

Further Reading